Pedagogical implications of the

lexical approach
Michael Lewis

Language has traditionally been divided into grammar and vocabulary.

Crudely, the former consisted of elements of the generative system of the
language and the latter was the stock of fixed nongenerative 'words'.
Recently, this analysis has been challenged and shown to be seriously
misguided from both strictly linguistic and pedagogical points of view
(Lewis, 1993; Nattinger & DeCarrico, 1992; Willis, 1990).
In fact, language consists broadly of four different kinds of lexical
items, the constituent 'chunks' of any language. Each chunk may be
placed on a generative spectrum between poles ranging from absolutely
fixed to very free. Although it is true that traditional vocabulary is usually
close to the fixed pole, and grammar structures are frequently close to the
free pole, this fact obscures the vastly more numerous and in many ways
more interesting items that occur nearer the middle of the spectrum.
These items may be 'words', or 'structures' in traditional language teach-
ing terms, but, as we will see, most typically they are lexical items of types
not recognised in most teaching material.
Lexical items are socially sanctioned independent units. These may be
individual words, or full sentences - institutionalised utterances - that
convey fixed social or pragmatic meaning within a given community.
This definition clearly entails that lexical items are dependent on agree-
ment within a particular social group; what is a lexical item in American
English may not be so in British English. Indeed, lexical items can be
peculiar to social groups of many kinds: geographical, professional, or
even family-, class- or age-based groups. Teenagers frequently use a
whole range of lexical items that leave their parents' generation at a loss.
Native speakers, in addition to words and grammar, have at their disposal
a repertoire of multiword items that are, for certain purposes, treated as
independent units. It is these items that have frequently been ignored.
Four fundamental types of lexical item may be identified:

Type 1 a. Word
b. Polywords
Type 2 Collocations

255
Downloaded from https://www.cambridge.org/core. University College Birmingham, on 16 Nov 2021 at 12:12:35, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139524643.018
 256 Michael Lewis

Type 3 Institutionalised utterances

Type 4 Sentence frames or heads
Such an analysis is not unique. There is some overlap between categories
and sometimes it is useful to consider a particular item as belonging to
different classes for different analytical purposes. This is particularly the
case when considering how such items may best be dealt with in the
classroom.

Words and polywords

Words have always been recognised as independent units. An utterance
may change its meaning by the change of a single word (Could you pass
my pen/calculator, please?) and single words may appear in speech or
writing as fully independent items (Stop, Sure!, Please.). This category of
lexical item has long been recognised in language teaching.
Polywords are but a small extension of this category. Although all
lexical items except words consist of more than one word, the term
'polyword' is restricted here to those (usually short) phrases that have a
degree of idiomaticity (by the way, on the other hand), and have usually
appeared in even quite simple dictionaries.

Collocations
Some pairs or groups of words co-occur with very high frequency, de-
pending perhaps on the text-type of the data. Most typically this feature
is associated with verb-noun and adjective-noun pairs (to raise capital, a
short-term strategy), but it can apply to word groups larger than pairs,
and to words from most grammatical categories. It will be noted that
although the structures of highly frequent verb patterns have formed, and
still form, a key element of most language courses, other highly frequent
word patterns - which is precisely what collocations are - have usually
been ignored or at best been seen as marginal to courses.
As Nattinger and DeCarrico (1992) pointed out, collocation is not
equally powerful in both directions - rancid strongly suggests the collo-
cate butter, whereas butter only weakly suggests rancid. Here the adjec-
tive is the 'key' word, carrying the most information content; most typ-
ically, the key word is a noun. This has important pedagogical
consequences.
Multiword collocations are linear sequences: to raise venture capital,
to dissolve the mixture in water, to make a determined effort. Typically,
they are 3-5 words in length and are primarily concerned with referential

Downloaded from https://www.cambridge.org/core. University College Birmingham, on 16 Nov 2021 at 12:12:35, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139524643.018
 Pedagogical implications of the lexical approach 257

content. There are obvious pedagogical advantages to ensuring both that

students are aware of the word partnerships (the nontechnical term I
prefer to 'collocation' for the classroom) and that when recording 'new
vocabulary', students record together and in sequence those words that
regularly occur in precisely that grouping. Too often students and
teachers look for and record only the new words in texts. This is to
misidentify the constituent chunks of the text in a way that is ped-
agogically unhelpful. The recognition, generation and effective recording
of collocations are essential elements of the Lexical Approach.

Institutionalised utterances
As the name suggests, these are more typical of the spoken than of the
written mode. They tend to express pragmatic rather than referential
meaning. They are all those chunks of language that are recalled as
wholes and of which much conversation is made. In their influential
paper, Pawley and Syder (1987) speculate that these items are vastly more
numerous than has ever been accepted. The chunks may be full sentences,
usable with no variation whatever but always with instantly identifiable
pragmatic meaning {Vllget it. It's nothing to do with me. There's a call for
you.), or sentence heads, which require another lexical item to provide a
complete utterance (If I were you, I'd wait). This last utterance is interest-
ing for it reveals the novelty within a lexical, rather than grammatical,
analysis.
Traditionally, If I were you, I'd wait is a 'conditional' or even a 'sub-
junctive'. If asked which two 'bits' make up the utterance, teachers al-
most invariably make the division between clauses. But this is incorrect.
We recognise that If I were you is ALWAYS followed by I'd, so the lexical
boundary between chunks is after I'd. Similar utterances are If I were
you, I'd/1 go, get one, leave the car at home. Notice that, under this
analysis, the utterance is simply the base form Go, Get one, Leave the car
at home preceded by a fixed chunk. Far from being a 'difficult structure',
it is as simple as it could be.
Traditional grammar has led teachers to believe that because language
items can be analysed in a particular way, it must be helpful to analyse
them in that way. But language can be analysed in many ways -
phonemes, syllables, morphemes or words. Different analyses are useful
for different purposes. Misidentifying the chunks of which language con-
sists has led to many pedagogical problems. In many cases the word is too
small a unit and the sentence is too large. Furthermore, those sentences
that are fully institutionalised utterances can be learned and used as
wholes, without analysis, thereby forming the basis, not the product, of
grammatical competence. This perception, extensively discussed by Nat-

Downloaded from https://www.cambridge.org/core. University College Birmingham, on 16 Nov 2021 at 12:12:35, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139524643.018
 258 Michael Lewis

tinger and DeCarrico, turns much of what has traditionally been called
'grammar teaching' on its head.
Institutionalised utterances, most importantly those that have the form
of a single fully grammaticalised sentence, should form a major element
in the language input provided for students. Such language is the basis of
natural language learning. Sadly, it is the very antithesis of the input
provided by many language courses, most notably those with (overt or
covert) behaviourist elements, which follow any version of the Present-
Practise-Produce paradigm. Willis makes a very similar point in his The
Lexical Syllabus. Fully contextualised (in the sense of occurring with
relevant co-text in a natural situation) institutionalised utterances repre-
sent the lexical item-type in which most language courses are most
deficient.
The precise problem that stimulated Pawley and Syder's original enqui-
ry was why it was that many utterances produced by nonnatives, while
not formally breaching the rules of well-formedness, were felt by native
speakers to be unnatural. It is helpful to recall that not all possible sen-
tences of English - i.e., those that are well formed according to the
patterns of the language - are actual or even probable utterances of the
language. Traditional grammar teaching concerned itself with all the pos-
sible sentences, believing that, in Chomsky's terms, performance derived
from competence, a concept essentially based on the total corpus of
possible sentences. In fact, acquisition appears to be based on induction
from natural utterances in the learner's input that are heard, read and (at
least partially) understood. Within this understanding it is clear that the
input should be biased heavily toward high-frequency utterances, most of
which will be fully or partially institutionalised.
Typically, institutionalised utterances are high on the 'Spectrum of
Idiomaticity'. This does not mean they belong to that group of rather
picturesque language items students and teachers frequently think of as
idioms (e.g., He threw in the towel; That's a whole new ball game) but
rather that the meaning of the whole is relatively nontransparent from the
meaning of the constituent words. This in turn is because many institu-
tionalised utterances contain a relatively high proportion of de-
lexicalised words, such as take, get and there, which do not, in them-
selves, carry much meaning. Typical utterances of this kind are I'll get it;
There's not a lot you can do about it; We just can't keep on like this.
Ironically, within the Lexical Approach, de-lexicalised words have a very
important part to play. The idiomatic nature of the wholes and the de-
lexicalised nature of the component words is reflected in this list of
archetypical utterances using 'II:

I'll give you a ring.

I'll be in touch.

Downloaded from https://www.cambridge.org/core. University College Birmingham, on 16 Nov 2021 at 12:12:35, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139524643.018
 Pedagogical implications of the lexical approach 259

I'll get back to you as soon as I can.

I'll be back in a minute.
I'll see what I can do.
You'll never get away with it.
It'll be all right.
It'll take time.
That'll do.
That'll be the day!
Nobody'll even notice.
There'll be hell to pay.
We'll see . . .
Although expressing instantly identifiable pragmatic meaning, and so of
immediate use to learners, such language forms a relatively small part of
the input in typical courses. Expressions like these deserve increased
attention, both because of their immediate usefulness and because they
provide input that is the basis for inductive acquisition of generalisable
'rules'. Additional emphasis on these institutionalised items must be ac-
companied by classroom strategies that make students more aware of
lexical items, and provide ample opportunities for them to practice such
language in the safety of the classroom.

Sentence frames and heads

These are to a large extent the written equivalent of institutionalised
utterances. They are those often large discourse features that allow us to
decode complex written text. The frustration of reading a student's essay
and thinking 'I know what you mean, but that's not the way to say (=
write) it', is most frequently caused by the student's failure to use this type
of lexical item. Some are comparatively short and easy (sequencers such
as secondly, . . . and finally)-, some are sentence heads serving similar
pragmatic purposes (We come now to a number of important reserva-
tions . . .); the largest represent those 'frames' that allow us to structure
long passages of text, usually written (e.g., the essay), but sometimes
spoken (the lecture or professional presentation). This type of lexical item
remains outside the field of competence even of many native speakers;
but it can be of great use to, for example, students of academic or profes-
sional English.
We have, then, four types of lexical item: the first two categories con-
cerned principally with referential meaning, the latter two with pragma-
tic meaning. A balanced language programme will have to take account
of all four types and, depending on the goals of the course, of the balance
between them. Broadly, the categories equate to the traditional ones of
Vocabulary (now Words and Collocations) and Function (now Institu-

Downloaded from https://www.cambridge.org/core. University College Birmingham, on 16 Nov 2021 at 12:12:35, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139524643.018
 260 Michael Lewis

tionalised Utterance and Sentence Frame). Interestingly, no lexical item-

type corresponds to such traditional structures as the present perfect,
conditionals or the passive. Grammar in this sense is radically de-
emphasised within the Lexical Approach.

Methodology
Krashen and Terrell (1983) contend that 'we acquire language by under-
standing messages'. If they are right, formal teaching may be of little
benefit to the learner. Indeed, it may be counterproductive in that it
frequently directs the learner's attention to individual words or gram-
matical structures that, as we have seen, are not the fundamental compo-
nents of the language. Although I am largely sympathetic to their view - 1
believe the structure of the language is acquired, as is the vast majority of
a learner's lexicon - 1 believe that activities that raise conscious awareness
of the lexical nature of language and its component chunks can aid
acquisition. The claim is a modest one - conscious awareness of what
constitutes a possible chunk provides learners with a tool that enables
them to process input more effectively.
It must be stressed that the primary purpose of the activities is
awareness-raising, rather than formal 'teaching'. Estimates of the mini-
mum functional vocabulary for the learner vary widely, depending on
technical factors such as what constitutes a 'word' or 'item', and different
views of functional competence. Nobody, however, estimates lower than
20,000 separate items (and most estimates are much larger). If each item
were formally taught, and took only 2 minutes of class time, and learning
were 100% efficient, nearly 700 hours would be required for this element
of the course alone. The truth is much closer to Krashen and Terrell's
position than many teachers find comfortable - the types and quality of
input must be carefully chosen, but we have less understanding of which
items will be acquired, and thus retained for future use, than is usually
acknowledged. Probably the best we can do is select input that is appro-
priate and encourage a low-anxiety atmosphere conducive to acquisition.
Within the lexical approach, less attention will be paid to individual
words and substantially less to traditional grammar structures; in con-
trast, much more time will be devoted to ensuring that students are aware
of the lexical items, particularly collocations, which carry much of the
(referential) meaning in written text, and institutionalised utterances,
which carry the meaning (in this case mostly pragmatic) of natural spo-
ken text. Many of the activities will be of the receptive, awareness-raising
kind. Teachers used to formal vocabulary teaching, using largely produc-
tive practice, need to make an important change of emphasis, learning
truly to value receptive practice.

Downloaded from https://www.cambridge.org/core. University College Birmingham, on 16 Nov 2021 at 12:12:35, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139524643.018
 Pedagogical implications of the lexical approach 261

Practices in the Lexical Approach

Productive practices are possible. In some ways the exercise types resem-
ble those of standard vocabulary or grammar teaching, although the lin-
guistic focus is different. In the sample exercises only a few examples are
given to illustrate the type of practice proposed; student materials would
need longer exercises with more examples.

Exercise type 1
In each of the following, one word does NOT make a strong word
partnership with the word in capitals; which is the odd word?
1. HIGH season price opinion spirits house time priority
2. MAIN point reason effect entrance speed road meal course
3. NEW experience job food potatoes baby situation year
4. LIGHT green lunch rain entertainment day work traffic
Note that there is little dispute about the word in examples 1 and 2, while
3 and 4 may give rise to discussion and doubt. This is intentional -
collocation is not a possible/impossible dichotomy, but likely/unlikely.
Although students (and teachers) prefer certainty to uncertainty, an ele-
ment of doubt is intrinsic to collocation. Care must be taken not to make
this type of exercise confusing, but totally unambiguous examples should
not be the sole target either.

Exercise type 2
Choose from these words four that make strong word partnerships in
business English with each of the verbs below.
bill presentation invoice discount debt lunch
deal calculation mistake service message expenses

PAY MAKE GIVE

Use some of the word partnerships to say something about your own job.

Again notice the use of the phrase strong word partnerships in the in-
struction. Most partnerships are possible; the exercise raises awareness of
strong or very likely partnerships.

Downloaded from https://www.cambridge.org/core. University College Birmingham, on 16 Nov 2021 at 12:12:35, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139524643.018
 262 Michael Lewis

Exercise type 3
Complete the table with five adjectives and five verbs that form strong
word partnerships with the noun VISIT.

Verb Adjective (Key word)

VISIT

Note that this simple array is most useful at the size proposed here;
smaller arrays are rarely sufficiently generative to be of pedagogical value,
while larger ones are confusing to students. The words that are included
in the array may be
a. Found from an accompanying source text.
b. Chosen from lists prepared in advance by the teacher.
c. Chosen from student suggestions.
In the last two cases the teacher must exercise careful judgement both of
the possibility of the collocation and, more importantly, of its value to the
student. 'Weak' adjectives (big, new, nice) rarely justify their place in the
recorded array. Collocates should be selected for inclusion on the basis of
their value in increasing the students' communicative power.

Exercise type 4
Fill in the middle column with an adjective that is opposite in meaning to
the word in the first column, but makes a correct word partnership with
the word in column three.
Column 1 Column 2 Column 3
helpful suggestion
efficient system
careful piece of work
safe choice

Downloaded from https://www.cambridge.org/core. University College Birmingham, on 16 Nov 2021 at 12:12:35, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139524643.018
 Pedagogical implications of the lexical approach 263

Column 1 Column 2 Column 3

light green
light suitcase
light rain
light work
Note that the idea of 'opposites' is invalid unless confined to contextual
or collocational opposites. Too often teaching oversimplifies this idea
unhelpfully.

Exercise type 5
His behaviour was very strange.
-Yes, very odd.
Her work is very careless.
-Yes, she's not careful at all.
Those roses are lovely.
-Yes, I like flowers around the house.
Everything seems to get more expensive all the time.
-Yes, especially food.
These examples, suggested by McCarthy (1991), show possible ways of
agreeing, using synonyms, or antonyms, and with more general or more
specific words. Recent work in discourse analysis suggests that 'lexical
agreement' of this kind is typical of natural conversation (Pearson, 1986;
Pomerantz, 1984). Oral exercises, resembling traditional grammar drills
but with a freer choice of lexically appropriate responses like those above,
provide excellent fluency practice.

Exercise type 6
The first part of each sentence in List 1 can be completed with the group
of endings given in List 2. Match the first parts with the endings.
List 1 List 2
1. I'm wondering a. to concentrate.
to understand it but I can't,
to remember where I put them.
2. I'm trying. b. quite pleased with myself.
a bit off colour.
more confident than I did.
3. I'm feeling c. what I can about it.
nothing yet!
the best I can.

Downloaded from https://www.cambridge.org/core. University College Birmingham, on 16 Nov 2021 at 12:12:35, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139524643.018
 264 Michael Lewis

4. I'm doing d. what we can do about it.

if it will make any difference,
if anyone else knows yet.
Can you think of a situation where you would say each of the twelve
sentences? Choose one of them and write a short dialogue so that one of
the people in the dialogue says exactly the sentence you have chosen.

Exercise type 7
Rearrange the following to make natural sentences.
1. It two about hours TAKES.
2. How it long will TAKE?
3. You I what don't say seriously TAKE.
4. I'll to a have just chance TAKE.
5. I'll a the from station taxi TAKE.
6. Don't to home books forget your TAKE.
7. I more any can't TAKE!
8. Nobody the of your parents can place TAKE.
When you go on holiday, do you take a lot of photos? Of what people or
places? What sort of souvenirs do you like to take home?
Are you the sort of person who:
a. takes responsibility b. takes advice c. takes risks?
Underline all the word partnerships with take in the examples.

Note that the rearrangement exercise is simplicity itself for the native
speaker but comparatively difficult for learners. It is essential that the
'answers' be institutionalised utterances chosen as input appropriate to
the particular group of learners. The 'conversation topics' are intended to
introduce a more affective quality, while still focusing on collocates of the
same de-lexicalised verb. The final underlining activity is important in
checking that students really have correctly identified the chunks. This
exercise type is useful for de-lexicalised verbs and nouns such as thing,
point, way. Whether the keyword should be highlighted or not is a matter
of taste (if it is not, many students simply do not notice that all the
examples contain the same word).

Exercise type 8
It must have been difficult.
-Difficult! It was absolutely impossible.
You must have been really fed up.
-Fed up! I was absolutely suicidal.

Downloaded from https://www.cambridge.org/core. University College Birmingham, on 16 Nov 2021 at 12:12:35, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139524643.018
 Pedagogical implications of the lexical approach 265

In pairs, make similar dialogues, using these expressions.

1. tired - exhausted 5. irritated - furious
2. interesting - fascinating 6. sure - certain
3. dangerous - treacherous 7. pleased - over the moon
4. surprised - astonished 8. expensive - a rip-off
When you speak your own language, are you the sort of person who says
7 was quite pleased, or do you prefer to say / was absolutely over the
moon}

This resembles a traditional structure practice, but now focus is on the

cross-turn lexical linking, which is typical of natural conversation. Syn-
onyms (contextual) and hyponyms frequently provide the raw material
for these exercises. Note also, the affective question. If students have to
learn many thousands of lexical items, it is important that the classes also
provide real opportunities for them to feel and to talk about their feelings.

Exercise type 9
Complete the following by adding one word. Only one word is possible in
each case. Make sure you know the equivalent expressions in your own
language.
1. I've no idea where my bag is - it's just disappeared into . . . air.
2. We weren't expecting her. She just turned up out of the . . .
3. It's . . . time something was done about it.
4. It's a difficult . . . of affairs. I'm not sure what to do.
5. It might have been quicker to take the car . . .all.
6. He might have changed his job by now, . . . all I know.
7. Oh, no! I've spilled coffee all . . . the place.
8. Revise carefully, but . . . all, don't panic during the exam.
The Lexical Approach widens the concept of 'idiom' so that attention
needs to be drawn to many natural phrases and whole sentences that were
previously excluded from course materials. As the examples above indi-
cate, both apparently meaning-carrying words (nos. 1-4) and de-
lexicalised words (nos. 5-8) need to be considered. The idea of 'equiv-
alent expressions' (rather than translations) is important in dealing with
institutionalised utterances.

Exercise type 10
Match each of these remarks with a response.
1. Would you like a cup of coffee?
2. Are you ready?

Downloaded from https://www.cambridge.org/core. University College Birmingham, on 16 Nov 2021 at 12:12:35, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139524643.018
 266 Michael Lewis

3. Have you got Jane's number there?

4. Everybody's worried about the situation.
5. Were you late last night?

a. I think so. Just a minute, I'll have a look.

b. No, we got there just in time.
c. Not just at the moment, thanks.
d. Oh, so it's not just me, then.
e. I will be in a moment. I'll just get my jacket.
This practice is in many ways typical of the Lexical Approach in that
while providing an activity for the student, the most important pedagogi-
cal element is that the practice material itself provides valuable input, in
this case the institutionalised replies.

Exercise type 11
Answer briefly - and honestly!

1. Think of three jobs you do at home - what are they and where in your
home do you do them?
2. Mention three public places or situations in your country where
smoking is not allowed.
3. Think of two or three important ideas you have had in connection
with your work or studies. Where did you get the ideas from?
4. Can you think of three ways you personally have made money without
working?

Note that the answers for this type of exercise must be true, thus avoiding
the random lexicalisation of practices of which Widdowson has com-
plained. Most of the answers will be short phrases, and many of these,
although not recognised as such by teachers, will be institutionalised: In
the kitchen, On public transport, From my parents, As Christmas or
birthday presents.
Of particular interest here is that the prepositional phrases are institu-
tionalised and, as such, precisely the kind of learned wholes that provide
the basis for, rather than being the product of, grammatical competence.

Exercise type 12
Some fixed expressions are made with more than one word. Complete
each of the phrases in List 1 with a word from List 2, then match the
expressions to the word in List 3 that is more common in written English.

Downloaded from https://www.cambridge.org/core. University College Birmingham, on 16 Nov 2021 at 12:12:35, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139524643.018
 Pedagogical implications of the lexical approach 267

List 1 List 2 List 3

1. by and then a. occasionally
2. in the long away b. immediately
3. every now and again c. repeatedly
4. time and large d. generally
5. straight run e. eventually

Exercise type 13
Some pairs or words can only be used in a particular order: Ladies and
gentlemen, but not * Gentlemen and ladies. Make pairs from these
words, find equivalents in your own language and use the English pairs in
natural sentences.
bread now rights down then ins back
forth outs butter out wrongs there here
Although polywords and pairs represent only a minor part of the total
lexicon, they should not be totally ignored.

Recording formats
Too much teacher training is precisely that - preparing teacher activity.
But it is the students who must learn and ultimately it is what they do that
facilitates learning, in more or less efficient ways. Having recognised this,
it is easier to see the importance of out-of-class activities.
Pre-class, the teachers' primary responsibility is the selection of high-
quality input. Equally if not more important, however, is preparing stu-
dents to make the best use of out-of-class activities such as independent
reading or watching TV.
Stevick has memorably remarked, 'If you want to forget something,
put it in a list', and all too often this is what students do with 'new
words' - list them, and forget them. It is essential to develop students'
skills in recording new lexical items they meet in or out of class in formats
that are a positive aid to retention. In each case the physical formatting
should mirror or highlight the real-world occurrence of the language.

Format 1
1. I'm so unfit. If I climb the stairs I'm completely out of ...
2. I couldn't steer the car at all. It was completely out of ...
3. I've repaid all the money I borrowed. At last I'm out of ...
4. There are none left, so you can't have one. You're out of ...
5. You certainly can't do that! It's out of ...

Downloaded from https://www.cambridge.org/core. University College Birmingham, on 16 Nov 2021 at 12:12:35, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139524643.018
 268 Michael Lewis

Note the alphabetical order of the answers, and the format, which high-
lights key expressions.

Format 2
You can HAVE lunch, a cup of tea, a plane to catch, a lot to do
PAY an invoice, bill, expenses
PAY FOR the tickets, lunch, a colleague
See also the 5 adjective-5 verb-key noun box already referred to (exercise
type 3).

Format 3
emphasise
point out
I should draw your attention to the fact that
remind you
explain
The precise formats vary, but the important point is that the recording
formats should be designed to suit the collocations, institutionalised ut-
terances, or sentence heads rather than simply listing items.

Additional exercises
As with all language practice, not all useful lexical exercises lend them-
selves to the generalisation that is clearly possible with the types listed
above. This is fortunate, for if the Lexical Approach is to work in the
classroom, as well as promote language proficiency, it must also provide
variety and enjoyment in class. Here are some activities that serve to
indicate the range of possibilities that are consistent with the Lexical
Approach.
1. Lexical Crosswords. The clues are lexical items, most typically utter-
ances, with missing words. The answers to the crossword are in fact
further clues to the utterances. The clues therefore represent valuable
input.
2. In pairs or small groups, write a story. Your story must:
a. Contain exactly 100 words - not one more or less.
b. Not use any English word (including a, the, is and other easy
words) more than once.
Once you have discovered some simple guidelines, this activity ensures
lexically dense stories. The most effective tend to use incomplete sen-
tences and 'de-grammaticalised dialogue-type' exclamations.

Downloaded from https://www.cambridge.org/core. University College Birmingham, on 16 Nov 2021 at 12:12:35, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139524643.018
 Pedagogical implications of the lexical approach 269

3. The teacher reads (or plays on tape) a story, chosen to include, for
example, a number of expressions using a de-lexicalised verb such as
get, keep. Students:
a. Raise a hand (silently) each time they notice an expression contain-
ing the keyword (which keeps all involved and helps weaker
students).
b. Write down the expressions containing the keyword.
4. Jigsaw Dialogues. The lines of a dialogue are rearranged, either as a
textbook exercise, or on strips of paper so that students can physically
rearrange them on the table or walk around the room sorting them-
selves into the correct order. Students then read the dialogue in the
correct order. It is important that the dialogue be written so that many
of its turns are, or contain, institutionalised utterances.
5. Lexical Dominoes. A set of cards is prepared so each card is divided
into like dominoes; on one half is written a noun, on the other a verb.
Students draw a group of cards - say, 7 each - and take turns to add
dominoes in the usual way. The two halves that touch must make a
strong word partnership.
This can also be done with noun-adjective cards. The words may be
chosen from a relatively small field such as the language for a particular
job, or, by using a number of de-lexicalised words, particularly verbs, a
more linguistically generative game can be constructed.
Many other activities with which teachers are familiar may be used
within the Lexical Approach. What is important is the attitude the
teacher takes toward the language of the input material. The two most
essential changes to the teacher's mind-set are a willingness to search for,
identify and direct attention toward the chunks of which all naturally
occurring language consists, and, methodologically, a ready acceptance
of the value of receptive, awareness-raising activities. Any classroom
procedure that respects these two key principles has a valid contribution
to make to a course informed by the Lexical Approach.

References

Krashen, S., $c Terrell, T. (1983). The natural approach: Language acquisition in

the classroom. Oxford: Oxford University Press.
Lewis, M. (1993). The lexical approach. Hove, England: Language Teaching
Publications.
McCarthy, M. (1991). Discourse analysis for language teachers. Cambridge:
Cambridge University Press.
Nattinger, J., & DeCarrico, J. (1992). Lexical phrases in language teaching.
Oxford: Oxford University Press.

Downloaded from https://www.cambridge.org/core. University College Birmingham, on 16 Nov 2021 at 12:12:35, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139524643.018
 270 Michael Lewis

Pawley, A., & Syder, F. (1987). Two puzzles for linguistic theory: Nativelike
selection and nativelike fluency. In J. Richards & R. Schmidt (Eds.), Lan-
guage and communication (pp. 191-226). London: Longman.
Pearson, E. (1986). Agreement/disagreement: An example of results of discourse
analysis applied to the oral English classroom. ITL Review of Applied Lin-
guistics, 74, 47-61.
Pomerantz, A. (1984). Agreeing and disagreeing with assessments: Some features
of preferred/dispreferred turn shapes. In J. Atkinson & J. Heritage (Eds.),
Structure of social action (pp. 57-101). Cambridge: Cambridge University
Press.
Richards, J., & Schmidt, R. (Eds.). (1987). Language and communication. Lon-
don: Longman.
Swan, M. (1984a). A critical look at the communicative approach. English Lan-
guage Teaching Journal 39(1), 2-12.
(1984b). A critical look at the communicative approach. English Language
Teaching Journal 39(2), 76-87.
Widdowson, H. (1991). Language awareness in language teaching education.
Paper presented at International TESOL, New York, March.
Willis, D. (1990). The lexical syllabus: A new approach to language teaching.
London: Collins ELT.

Downloaded from https://www.cambridge.org/core. University College Birmingham, on 16 Nov 2021 at 12:12:35, subject to the Cambridge Core
terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/10.1017/CBO9781139524643.018